Or is there such a thing as data with no action-related intentions at all? What would that look like? What is its value?
When I first schemed about this blog, I thought “sure, data is definitely collected that is not action-focused, action-oriented, or action-generating.” Data collected for the sake of data alone. Or data that somehow isn’t actionable once collected.
But now I’m not so sure.
Earlier this year I explored a section of Michael Quinn Patton’s “Utilization Focused Evaluation” with a few colleagues and I find that I am still pondering it in relation to the question of what exactly is ‘actionable data’ (and as a result, what isn’t).
This section of Patton’s book covers what is essentially a variety of types of uses for data (actions, if you agree that we can extend ). I find the process and enlightenment uses particularly intriguing in light of defining actionable (and non actionable) data.
My understanding of process use, per Patton’s discussion, is essentially that the evaluation (or assessment) process in and of itself results in changes to the process or the program being studied.
I find enlightenment use to have a broader wingspan, conceptually. It covers the growth of ideas and understanding that result from evaluation, it can also refer to changes in thinking such as changes in assumptions or priorities. — this to me is really about learning. Learning as individuals and as communities.
Let’s assume for a moment that we’re talking about ‘good’ data — i.e. not bad data (that’s the subject for an entirely different post). Good data, for now, means that it’s data that has a reasonable amount of relevance, reliability, validity, and quality with regard to whatever it is we’re trying to learn about.
Now, it seems to me that given the uses described in Patton, all (good) data has the potential to be useful, in at least ‘enlightenment’ and ‘process’ areas if nothing else. Doesn’t data collected for information’s sake, or even data which appears to not be actionable once collected, have the potential for enlightenment use? And isn’t this action in its own right? Albeit more intellectual than tangible, does that not move ‘us’ forward in some way?
For example, let’s consider data collected from our annual alumni survey that we realize only upon collection didn’t entirely capture what we wanted, or needed (and perhaps our needs have changed some as we want to delve deeper than we thought we did originally). We now know this thing about alumni, but only kinda, and as a result, we want to add or change questions in the survey for next year. This data doesn’t seem actionable in the sense that we can’t do much actively with it, but we have actually gained understanding about what it is we need to know as a result of collecting responses and discussing results; it’s purpose is in enlightenment use. And stay with me here… we’re ‘closing the loop,’ taking action to revise the survey tool as a result of our assessment of that tool, and a crucial player in that assessment was the data that didn’t seem actionable. Clear as mud?
Now, this is not to be confused with data that should or could be actionable but fails to be due to simple lack of use — examples of that sit on shelves of business, non-profit, and government offices throughout the US (and likely the world), or worse yet, misuse (cue examples of shoddy poor political polling used throughout the news media).
So what is it that makes data actionable then? Does it lie in the data itself? The way it is collected? Analyzed? Disseminated? The context/environment in which the data is collected, analyzed, & disseminated? In some sort of special concoction where all these aspects are blended perfectly together? Does it lie solely in intentions (an input, if we want to get meta about this) and tangible use (an outcome) and all the stuff in the middle, the actual collection, analysis, etc., is just fluff? or is that stuff in the middle more brick and mortar? How important is the role of context/environment?
I feel a post about the importance of context coming on (and flashbacks to my graduate coursework in policy — does this remind anyone else of ‘policy windows’?)…. Not to mention a post about ‘bad’ data.